2. Impact of Reimbursement Decisions on Innovation in Genetic Testing
Elaine Lyon, PhD
Medical Director, Molecular Genetics, Co-Medical Director, Pharmacogenetics
Associate Professor of Pathology
University of Utah School of Medicine
President-elect, Association for Molecular Pathology

3. Objectives
Discuss the molecular pathology tier coding and issues with reimbursement
Describe implications of reimbursement decisions for complex genetic testing
List approaches to demonstrate utility of genetic testing

4. Molecular Pathology Codes General principles
Accepted clinical practice
Multiple labs
Typical test
Increased transparency
Tier 1 – frequently ordered
Tier 2
Unlisted

5. Tier 1 Examples: Transparent Inherited disease Somatic diseases
81206: BCR/ABL1 (t(9:22)) (eg, chronic myelogenous leukemia translocation analysis; major breakpoint, qualitative or quantitative
81207: > minor breakpoint, qualitative of quantitative
81208: > other breakpoint, qualitative or quantitative
81220: CFTR (cystic fibrosis transmembrane conductance regulator) (eg, cystic fibrosis) gene analysis, common variants (eg ACMG/ACOG guidelines)
(when intron 8 poly-T analysis is performed in conjuction with in a R117H positive patient, do not report 81224)
81221: > known familial variants
81222: > duplication/deletion variants
81223: >full gene sequence
81224: >intron 8 poly-T analysis (eg, male infertility)
Examples:
Transparent
Inherited disease
30% of test menu
87% of test volume
Somatic diseases
41% of test menu
63% of test volume
Represents ‘mature’ tests
Often simpler, less expensive tests

6. Tier 2 Examples: Not complete transparent
Amount of work
Laboratories do not ‘self-assign’
Inherited disease
32% of our test menu
6% of test volume (genetic),
Somatic disease
38% of test volume
20% of test volume
Costs reflected in level
Codes may move to Tier 1 as become routine
81401: Level 2 (eg, 2-10 SNPS, 1 methylated variant or 1 somatic variant (typically using nonsequencing target variant analysis) or detection of a dynamic mutation disorder/triplet repeat)
ABL (c-abl oncogene I, receptor tyrosine kinase) (eg, acquired imatinib resistance). T315I variant
81407: Level 8 (eg, analysis of exons by DNA sequence analysis, mutation scanning or duplication/deletion variatns of >50 exons, sequence analysis of multiple genes on one platform)
> F8 (coagulation facotr VIII) (eg hemophilia A), full gene sequence

7. Unlisted: 81479 Examples: Inherited disease Somatic disease
Blood antigens
IL28B
Translocations
BCL-2/JH (t(14;18), API2-MALT1 t(11;18), SYT-SSX, t(X;18),
Rare genetic diseases
mitochondrial genome
Exome
Inherited disease
31% of test menu
4% of test volume
Somatic disease
7% of test menu
14% of test volume
Percentages may differ significantly for specialty laboratories

8. Next Generation Sequencing
Codes not available
Oncology/inherited/infectious disease
Gene panels, exomes, genomes
Intepretation
Exomes: used as gene panel, 2nd analysis of exome if 1st analysis not informative
Genomes: Analzyed as panel or exome, 2nd analysis of genome if 1st analysis not informative
Symptom-guided analysis vs incidental findings
Coding for re-analysis/re-intepretation?
IMPACT: Delayed tests from being brought on-line

9. NGS Examples
Non-invasive prenatal diagnosis of aneuploidies (cell free fetal DNA)
Multi-gene panels for heritable disorders
Exome/genome for rare genetic disorders
Multi-gene/multiple mutation for oncology
Clonality assessment in lymphomas
Genome/exome of a neoplasm
Microbiome

10. Cancer Panel 50 genes 2855 Mutations codes:
81210 (BRAF) $180.60
81235 (EGFR) $331.50
81245 (FLT3) $167.17
81270 (JAK2) $126.00
81275 (KRAS, ex 12-13) $179.20
81310 (NPM1) $249.01
81402 (MPL) tier 2 not listed
81403 x5 (ABL1, HRAS, IDH1, IDH2, KRAS ex 3)
81404 x6 (FGFR2, FGFR3, KIT, NRAS, PDGFRA, RET)
81405 (TP53)
81479 x1 (the rest) OR ALL
Impact: uncertain reimbursement
ABL1
AKT1
ALK
APC
ATM
BRAF
CDH1
CDKN2A
CSF1R
CTNNB1
EGFR
ERBB2
ERBB4
EZH2
FBXW7
FGFR1
FGFR2
FGFR3
FLT3
GNA11
GNAQ
GNAS
HNF1A
HRAS
IDH1
IDH2
JAK2
JAK3
KDR
KIT
KRAS
MET
MLH1
MPL
NOTCH1
NPM1
NRAS
PDGFRA
PIK3CA
PTEN
PTPN11
RB1
RET
SMAD4
SMARCB1
SMO
SRC
STK11
TP53
VHL

11. Marfan Single Gene Assay
66 exons
Mutation Positive (10% positivity rate)
Includes known pathogenic and suspected pathogenic
56%: diagnosis based on clinical phenotype
44%: suspected diagnosis of Marfan disease
 4% Variants of Uncertain Clinical Significance
64%: Suspected diagnosis of Marfan
37%: Diagnosis based on clinical phenotype
p. Arg545Cys

12. Ehler Danlos Syndrome Type IV
Marfan syndrome
Tall stature
Arachnodactyly
Hypermobile joints
Scoliosis
Aortic aneurysm
Learning disability
Positive family history, sudden death in a close relative
Loeys-Dietz Syndrome
Arterial tortuosity
Hypertelorism 

Bifid (split) or broad uvula
Aneurysms
Scoliosis
Positive family history, sudden death in a close relative
Morris et al., 2011 Circulation
Arterial Tortuosity
Tortuosity, elongation, and aneuryms of major arteries and the aorta
Aortic stenosis, pulmonary artery or pulmonary valve
Hypertelorism
Hypermobile joints
Arachnodactyly
Scoliosis
Hyperextensible skin
Positive family history, sudden death in a close relative
Ehler Danlos Syndrome Type IV
Aneurysm
Thin, translucent skin
Extensive bruising
Hypermobility
Clubfoot
Spontaneous pneumothorax or haemothorax
Positive family history, sudden death in a close relative
When we look at the clinical findings of these disorders, Marfan syndrome, Loeys-Dietz syndrome, Ehler danlos type IV, arterial turtuosity, we see this group of disorders have very similar findings. These patients may come with aortic aneurysm, tall stature and some skeletal findings. It will be difficult to pin point which disorder patient may have. As you have seen in this slide, Positive family history is also one of the common finding with sadly sudden death in a close relative. Finding the causative gene/mutation is very important for the family members as well.
Cummings et al., 1998 JBJS
Courtesy of Dr. Pinar Bayrak-Toydemir

13. Utility of NGS Gene Panel Assays
Overlapping phenotypes or genetically heterogeneous disorders:
Marfan Syndrome: most common (1:5000)
FBN1 is a large gene (66 exons)
expensive & time consuming to sequence by Sanger
Loeys-Dietz syndrome (LDS) and Ehlers-Danlos syndrome (EDS) type IV are clinically related to Marfan syndrome & difficult to distinguish clinically.
Provide accurate diagnosis
Useful for patients
Eliminate repeated or non-essential medical evaluations
Eliminates time and cost of sequential sequencing of genes
Useful for family members:
Mutation identification to prevent at-risk family members from complications
AS I mentioned at the beginning gene panels are ideal for overlapping phenotypes and less stressful and cost effective.
Courtesy of Dr. Pinar Bayrak-Toydemir

14. Association for Molecular Pathology GSP Codes
Proposal to address CPT coding for Genomic Sequencing Procedures (March 2013)
Association for Molecular Pathology Economic Affairs Committee
Jeff Kant, former chair, Jan Nowak, co‐chair, Aaron Bossler, co‐chair, Dara Aisner, Sam Caughron, Jill Hagenkord, Roger Klein, Elaine Lyon, Paul Raslavicus, Linda Sabatini, Michelle Schoonmaker, Ester Stein, Kathryn Tynan

15. Considerations Transparency Clinical Utility Unique features of GPS
Assay many genes simultaneously
Re-analysis (re-query) of data
Breadth of coverage
Depth of coverage

16. Reimbursement Unlisted code not equivalent to denial or under payment
CPT codes do not ensure reimbursement
Received notification that some pharmacogenetic applications (cytochrome enzymes) will not be covered
Others (CFTR, Long QT) require pre-authorization

17. Clinical Validity/Utility
Mendelian diseases: gene mutations cause disease
Association: risk for common diseases
Clinical utility
Payor definition
Change management and improve outcomes

18. Define Clinical Utility
Physician
Change management and improve outcomes
Appropriate diagnosis stops diagnostic odyssey
Patient
Prognosis
Personal utility
Family and life planning
Inheritance pattern and recurrence risk
Family
Utility for siblings/children

19. Case Affected male with developmental delay Cytogenomic microarray
X28q loss – known pathogenic
Recommendation:
Test mother to determine if de novo
Affects recurrence risk
Testing not covered for mother
Son is Medicaid patient
Microarray is very expensive
Requested ‘no charge’ testing of qPCR
Impact: Risk to 2nd child remains unknown unless laboratory provides free testing

20. Establishing Clinical Utility
Randomized prospective controlled studies
Retrospective
Issues with
Rare inherited diseases
Long duration
Diagnostics treated as pharmaceuticals or devices
Safe and effective?
Accurate and precise!

21. Mitochondrial Nuclear Genes
Clinically heterogeneous group of disorders … of dysfunction of the mitochondrial respiratory chain.
Common clinical features:
ptosis, external ophthalmoplegia, proximal myopathy and exercise intolerance, cardiomyopathy, sensorineural deafness, optic atrophy, pigmentary retinopathy, and diabetes mellitus.
Common central nervous system findings:
fluctuating encephalopathy, seizures, dementia, migraine, stroke-like episodes, ataxia, and spasticity.
Management is largely supportive
may include early diagnosis and treatment of diabetes mellitus, cardiac pacing, ptosis correction, and intraocular lens replacement for cataracts. Individuals with complex I and/or complex II deficiency may benefit from oral administration of riboflavin
GeneReviews:
Conservative prevalence estimate of all mitochondrial diseases is 1:8500
“mitochondrial nuclear gene tests do not support the required clinical utility for the established Medicare benefit category and are statutorily excluded tests”. 
Impact: Not reimbursed – can we continue to offer the test?

22. Design Difficulties Enough statistical power:
Examples:
CYP2D6 combining heterozygous and homozygous to get enough numbers
Mitochondrial nuclear genes: >100 genes
Changing Standard of Care
Example: Warfarin
Lyon et al. Laboratory testing of CYP2D6 alleles in relation to tamoxifen therapy .Genet Med 2012:14(12):990–1000
Endoxifen plasma concentration

23. EGAPP Evaluation for Genomic Applications and Practice and Prevention
Common conclusion
Insufficient evidence
… found insufficient evidence to support a recommendation for or against use of CYP450 testing in adults beginning SSRI treatment for non-psychotic depression.
In the absence of supporting evidence …., EGAPP discourages use of CYP450 testing for patients beginning SSRI treatment until further clinical trials are completed.
Taken as “Evidence Against”

24. Evidence-Based Reviews
Genetic Test
Reviewed by
Conclusion
Thrombophilia
AHRQ/EGAPP
No direct evidence
HER2
AHRQ
Weak evidence
Gene expression/breast cancer
High quality retrospective for Oncotype DX
UGT1A1
EGAPP
Insufficient evidence
HNPCC
Limited evidence for benefits to family members
CYP450 – non-psychotic depression
Paucity of good quality evidence (SSRI)
Ovarian cancer
No evidence
Adapted from Leonard, IOM workshop presentation, Nov 17, 2010: In: Generating Evidence for Genomic Diagnostic Test Development: Workshop Summary: National Academy of Sciences.

25. Coverage with Evidence Development
Class II code
Example: Warfarin
CYP2C9 (*2, *3)
VKORC1 Promoter mutation (-1639G>A)
Genetics explains 30-50% of warfarin dose variability
Clinical features explain 20-30%
>20% of dose variability remains unexplained
The gene-dose relationship is clear & convincing
People with variants are at risk for over-dosing
Improved outcomes not proven
Test volume dropped after decision, never has grown

26. Circular Problem Marker utility poorly valued Lower evidence
Not reimbursed
Lower funding/lack of interest
Lack of clinical trials
Lower evidence
Adapted from: Generating Evidence for Genomic Diagnostic Test Development: Workshop Summary: National Academy of Sciences.

27. Models Include Types: Biological relationships
Treatment, diagnosis, disease penetrance
Personal/Family utility
Types:
Oncology
Inherited diseases

28. Examples of Models Oncology
Prognostic, predictive, markers, companion diagnostics
Chain of evidences
Demonstrate usefulness in one or multiple cancer/specimen types?
Define “ supportive” and ‘adequate” evidence
JNCCN 9(S5) 2011 NCCN Task Force Report; Evaluating the Clinical Utility of Tumor Markers in Oncology
Generating Evidence for Genomic Diagnostic Test Development: Workshop Summary: National Academy of Sciences

29. Inherited Disease Autosomal dominant Available treatment
Hereditary hemorrhagic telangiectasia
Galactosemia
No available treatment
Huntington disease
Autosomal recessive
Cystic fibrosis testing for adults
Carrier status
Atypical CF-related symptoms
X-linked
Fragile X

30. Establishing Utility
Molecular pathologists/clinical laboratorians, clinicians, health economists
Underpowered or partial data modeled for useful information
Will require models for different scenarios

31. Regulation and Reimbursement
Find a standard that can be accepted by both regulations and reimbursements
Impact: Priceless

32. Summary Codes needed but not sufficient for reimbursement
Impact of reimbursement challenges
Uncertainty for Laboratories
Insufficient or lack of reimbursement
Delay, discontinue or not develop tests
Definition of Utility needs to expand
Family members to receive appropriate testing
Need different models for utility

33. Thanks to ARUP Laboratories Association for Molecular Pathology
Clinical and R&D Molecular Genetics/Genomics
Association for Molecular Pathology
Professional Relations Committees
Clinical Practice
Economic Affairs

34. 2013